Across Dimensions: Developing 2D and 3D Human iPSC-Based Models of Fragile X Syndrome

Abstract

Fragile X syndrome (FXS) is the most common inherited cause of intellectual disability and autism spectrum disorder. FXS is caused by a cytosine-guanine-guanine (CGG) trinucleotide repeat expansion in the untranslated region of the FMR1 gene leading to the functional loss of the gene's protein product FMRP. Various animal models of FXS have provided substantial knowledge about the disorder. However, critical limitations exist in replicating the pathophysiological mechanisms. Human induced pluripotent stem cells (hiPSCs) provide a unique means of studying the features and processes of both normal and abnormal human neurodevelopment in large sample quantities in a controlled setting. Human iPSC-based models of FXS have offered a better understanding of FXS pathophysiology specific to humans. This review summarizes studies that have used hiPSC-based two-dimensional cellular models of FXS to reproduce the pathology, examine altered gene expression and translation, determine the functions and targets of FMRP, characterize the neurodevelopmental phenotypes and electrophysiological features, and, finally, to reactivate FMR1. We also provide an overview of the most recent studies using three-dimensional human brain organoids of FXS and end with a discussion of current limitations and future directions for FXS research using hiPSCs.

Keywords: FMRP; fragile X syndrome; iPSC; organoids.

Figure 1

Summary of major findings from 2D FXS−iPSC−derived models. (A) Studies that have utilized 2D hiPSC-derived models of FXS have found that the fundamental molecular mechanisms of FXS pathology is retained in iPSCs derived from primary cells of FXS patients (FXS-iPSCs) [46,47]. (B) FXS-iPSCs and derived neural cells showed altered gene expression and translation compared to controls [21,52,53,54,55,56,57]. (C) Studies of targets and functions of FMRP showed that there are cell-type-specific [23] and function-specific [58] targets of FMRP [23,58]. (D) Neurodevelopmental abnormalities, including increased NPC proliferation [21,23,53] and altered neurite growth [47,48,53,55,59], were characterized in iPSC-derived models of FXS. (E) FXS-iPSC-derived neural cells exhibited electrophysiological abnormalities [24,59,60,61,62,63]. (F) Pharmacological rescue [64,65,66,67,68,69] and gene editing [70,71,72,73] are two major methods of FMR1 reactivation that have been explored.

Figure 2

Summary of major findings from 3D FXS−iPSC−derived models. (A) FXS-iPSC-derived forebrain organoids exhibited altered NPC proliferation and neural differentiation [21,22]. In particular, increased NPC proliferation was observed at differentiation day 28 [21], while reduced NPC proliferation and accelerated neural differentiation were observed at day 56 [22], suggesting a developmental-stage-specific alteration in NPC proliferation rate. (B) Increased gliogenesis was observed in FMRP-KO cerebral organoids [24]. (C) FXS-iPSC-derived forebrain organoids showed increased synapse formation and hyperexcitability at differentiation day 56 [22]. (D) Transcriptome analysis revealed alterations in gene expression profile and cell-type-specific developmental trajectory in FXS-iPSC-derived forebrain organoids [21,22]. (E) Using human forebrain organoid models, a large number of human-specific FMRP targets were identified via eCLIP-seq [22]. (F) PI3K inhibitors were able to rescue FXS phenotypes in forebrain organoids [22].

Figure 3

Summary of future directions for FXS-iPSC research. Future studies of human iPSC-based models of FXS should aim to (A) improve reproducibility, (B) enhance the representation of neural network structure and organization through the incorporation of various glial cell types and use of assembloids, and (C) provide a comprehensive examination of neurodevelopment from early to late stages by developing models that allow for prolonged culture, such as vascularized and sliced organoid models, as well as by corroborating results through a combination of 2D and 3D models.